Genome-wide methods for mapping transcription factor binding sites in vivo, by ChIP-seq and ChIP-chip methods, have led to an explosion of information about the deployment of regulatory factors across the genome in living cells. The parts list for developmental systems biology now includes not only regulatory factors but also many candidate cis-regulatory target sites. Still, a major problem for understanding transcriptional regulation in development is how to match transcription factor site occupancy with sites of actual functional impact. Static transcription factor binding snapshots derived fro cell lines or terminally differentiated states may indicate sites where a factor is working in thos cells. However, in fine-scale examination of actual developmental systems, dynamic changes in factor occupancy occur in parallel with changes in RNA expression. The functional impact of factor binding at many sites is obscure, and many sites of factor occupancy may be superfluous for transcriptional control. This underlines the urgency of finding better methods to predict functionality of transcription factor engagement at specific sites. This project is based on a global strategy for elucidating relationships between genome-wide factor binding and transcriptional function, in a well-established developmental system in which multipotent blood-cell precursors become committed to develop as T cells. Our previous work has defined the transitions in this process at both cellular and molecular levels, including a recent global analysis of stepwise transcriptome changes, changes in transcription factor gene expression, local chromatin marking changes, and changes in magnitude and sites of binding occupancies across the genome by specific transcription factors. The multistage information about regulatory components of this process makes it possible to analyze transcription factor action dynamically, not only in terms of where binding is seen, but also in terms of the changes over time in binding occupancy at each site, which can then be compared to expression changes in the linked genes. To evaluate whether the transcription factor binding changes actually cause these gene expression changes, functional perturbations are required. Here, we propose to apply a systematic program of fast, stage-specific perturbations to determine the mechanisms of action of two essential transcription factors, PU.1 and Runx1. An important set of tools will be obligate repressor constructs used as competitors for endogenous factors to distinguish direct vs. indirect activation and repression. Quick gain and loss of function can thus be applied at particular stages and the effects analyzed by RNA-seq to identify affected genes system-wide, even if effects of a factor change magnitude or sign between developmental stages. The sites and site combinations thus suggested to have functional impact will be examined for distinctive co-clustered sites, histone modification rules, and transcription factor exchange rates. These features will then be tested for generality by functional assays of new, specifically predicted cis regulatory site modules within the gene network.